Generative AI Strategy for a Car Company: In-Vehicle and Internal LLM Use Cases
Company: Google
Role: Product Manager
Category: Product Design & Strategy
Difficulty: medium
Interview Round: Other
You are a product manager at a large car manufacturer. Should the company use generative AI or large language models in the cars it sells and in the way it works internally? Make a recommendation and explain how you reached it.
```hint Judge each use by the cost of a wrong answer
List the places where customers, dealers and employees deal with language-heavy or document-heavy work, then weigh each one's value against what happens when the model gets it wrong. A mistake in a back-office summary and a mistake inside a moving car are very different.
```
### Clarifying Questions
- Is this a mass-market or a premium brand, and do its newer cars already receive software updates over the air?
- Should I cover both in-vehicle features and internal workflows such as engineering, service, sales and support, or focus on one side?
- Can the company use models from a cloud provider, or are there limits on sending vehicle and customer data off the car or outside the company?
- Is leadership mainly after new software revenue, lower costs, or a better customer experience?
### What a Strong Answer Covers
- A clear recommendation that says where to use generative AI and where not to, rather than a list of ideas
- Opportunities across in-car products, customer channels and internal workflows, ranked by value, feasibility and cost of error
- Firm limits for driving safety, driver distraction, liability, privacy and regulation
- A phased plan with first pilots, a build-versus-partner choice, and success metrics with guardrails
- Practical constraints: in-car compute and connectivity, inference cost across a fleet, and cars that stay in service far longer than a model generation
### Follow-up Questions
- The CEO wants an LLM voice assistant in every new model next year. What would you need to see before agreeing, and what would you ship if it is not ready?
- How would you stop an in-car assistant from giving a wrong answer about the vehicle itself, such as what a warning light means?
- A rival automaker launches a chatbot in its cars to strong press coverage. Does that change your plan?
Overview: A product strategy question asking whether a large car manufacturer should adopt generative AI and large language models in its vehicles and internal workflows. It tests how a candidate maps use cases, weighs value against the cost of wrong outputs and driver safety, and turns the analysis into a phased, measurable recommendation.